Abstract | | |
The new predictive formula generated during the study of Modification of Diet in Renal Disease (MDRD) to estimate the glomerular filtration rate in chronic kidney disease (CKD) patients was found to be superior to existing predictive formulas in all races including black Americans. We had previously published a study evaluating and comparing 5 predictive formulas and their applicability in Nigerian CKD patients and normal subjects. The existing data from this study were re-analyzed and the 5 previous formulas compared with the MDRD formula. All the predictive formulas including the MDRD formula correlated significantly with measured creatinine clearance in CKD subjects and controls. Correlation Coefficient, (r) ranged between 0.908-0.968 and Coefficient of Determination, (r^{ 2} ), ranged between 0.826-0.936. There was also good correlation between the measured and predicted CrCl in healthy state, though the r and r^{ 2} values were weaker (0.718-0.957) and (0.516-0.916). Specifically, MDRD formula was only superior to Jelliffe and Gates and not so to Cockcroft and Gault, Hull, and Mawer equations in CRF. MDRD formula yielded r= 0.93 and r^{ 2} = 0.86 and the values for Cockcroft and Gault, Hull and Mawer ranged between 0.96-0.97 and 0.93-0.94 respectively. In conclusion, MDRD formula, though useful and applicable was not superior to existing formulas. Cockcroft and Gault equation can still be used due to the ease of recall and its high correlation coefficient in health and disease states. **Keywords:** Creatinine clearance, Glomerular filtration rate, Chronic kidney disease, Predictive formulas
**How to cite this article:** Abefe SA, Abiola AF, Olubunmi AA, Adewale A. Utility of predicted creatinine clearance using MDRD formula compared with other predictive formulas in Nigerian patients. Saudi J Kidney Dis Transpl 2009;20:86-90 |
**How to cite this URL:** Abefe SA, Abiola AF, Olubunmi AA, Adewale A. Utility of predicted creatinine clearance using MDRD formula compared with other predictive formulas in Nigerian patients. Saudi J Kidney Dis Transpl [serial online] 2009 [cited 2019 Sep 22];20:86-90. Available from: http://www.sjkdt.org/text.asp?2009/20/1/86/44711 |
Introduction | | |
Glomerular filtration rate (GFR) is an important index of measurement of clinical course of renal disease because the rate of glomerular filtration generally is believed to be the overall index of renal function in health and disease.^{ [1]} Currently, the classification of chronic kidney disease (CKD) into 5 stages (by National Kidney Foundation) relied heavily on the different values of GFR.^{ [2]} Therefore, GFR is a useful and routine tool in the management of chronic renal failure patients.
Accurate determination of GFR using endogenous creatinine clearance in clinical practice is beset with a number of problems. These problems relate to the difficulty in sample collection, performance of the test, inconvenience to patients, waste of work time, and use and cost of concomitant drugs. In addition, incomplete urine collection sometimes result in imprecise estimation of GFR.^{ [3]} Other more accurate modalities for assessing GFR are either unavailable or very expensive and beyond the reach of most patients particularly in the developing world.
Several formulas have been developed for rapid and reliable determination of creatinine clearance that are comparable to measured creatinine clearance. MDRD formula is the most recent addition.^{ [4]} In our earlier publication, we compared 5 existing formulas with traditionally measured 24 hours creatinine clearance in 34 chronic renal failure patients and 32 normal individuals.^{ [5]} We concluded that the predictive formulas were satisfactory both in health and disease states and that Cockcroft and Gault formula^{ [6]} had the highest sensitivity and specificity and superior of the five assessed formulas. However, since the introduction of MDRD formula to predict CrCl from the Modification of Diet in Renal Disease (MDRD) study, several studies have dcumented the superiority of this formula over the existing ones.^{ [7],[8],[9]} The present study re-analyzes the data from our previous study using current MDRD formula to determine its utility in both health and disease states and compare it with previous five formulas namely, Jelliffe,^{ [10],[11] } Mawer,^{ [12]} Cockcroft and Gault,^{ [6]} Hull^{ [13]} and Gates^{ [14]} in an homogenous African population.
Patients and Methods | | |
The study was carried out in 32 healthy subjects and 34 patients with established CKD having serum creatinine consistently above 177 µmol/L. Only patients passing at least 500 mL of urine in 24 hours and who had not been previously dialyzed were recruited to the study. None of the study patients and normal controls was on any of the following drugs: salicylate, co-trimoxazole, trimethoprim, cimetidine, or probenecid. Patients with massive edema, jaundice, liver disease, and ketosis were excluded.
All patients were admitted into the Renal Ward of the hospital for supervised 24-hour urine collection. At the end of urine collection and in fasting state, 10 mL of venous blood was taken into lithium heparin specimen bottle for chemistry. Urine volume was also determined, and an aliquot was taken for chemistry. Patient's weight and age were recorded. Similarly, normal subjects who consisted of doctors, nurses and laboratory scientists, underwent similar procedure. Blood and urine creatinine estimation were done using diacetylmonoxime and kinetic Jaffe method.
Statistical methods | | |
The statistical package used to analyze the data was SPSS for Windows version 11. The relationship between measured and predicted creatinine clearance was evaluated by linear regression analysis and the intrinsic strength of the predicted CrCl against measured CrCl was determined by Coefficient of Determination (r^{ 2} ). Comparisons of the prediction formulas were performed using 1-R^{ 2} (which is degree of variance of each of the formulas from predicting the actual GFR using endogenous creatinine clearance as gold standard) in both established CKD and normal controls.
Results | | |
The mean age for the patients and healthy controls was 34.97 ± 11.2 and 34.13 ± 10.0 years, respectively. The means of serum creatinine levels for male and female patients and healthy controls was 682.0 ± 354.5 µmol/L and 866.7 ± 433.5 µmol/L, p= 0.189. The mean serum creatinine level for the normal controls was 85.3 ± 33.3 µmol/L. The mean urinary creatinine excretion levels for male and female patients were 7636.0 ± 3889.1 µmol/24 hours and 7329.6 ± 4084.9 µmol/24 hours, p= 0.83.
[Table 1] shows the regression parameters between the measured CrCl and predicted formulas for both patients and normal subjects in addition to the percentage of variance for each of the formulas from predicted GFR.
[Table 2] shows these parameters with their slope and intercept and P-value in both established CRF and normal controls to determine the suitability and superiority of the formulas.
[Figure 1],[Figure 2],[Figure 3],[Figure 4] show the linear regression graphs of Cockcroft and Gault and MDRD in established CKD and in health.
Discussion | | |
GFR is an important index for assessment of the clinical course of renal disease, and forms the basis for classification of CKD into different stages.^{ [2]} The stages that are derived from GFR closely reflect other abnormalities of renal function apart from excretory function and thus equip nephrologists with a necessary tool to establish management strategies. In our earlier publication, we examined the usefulness of five different predictive formulas and compared them with the GFR derived from traditional 24-hour urine endogenous creatinine clearance and concluded that Cockcroft and Gault formula in contrast to Jelliffe, Mawer, Hull and Gate formulas offered the best result in both health and disease states. Therefore, it was concluded that Cockcroft and Gault formula could be used in place of time consuming and laborious 24-hour urine endogenous creatinine clearance determination,^{ [5]} in addition to the recent MDRD formula, which was endorsed by some authors.^{ [7],[8],[9]} In this study, the MDRD formula was not found superior to Cockcroft and Gault formula in both healthy controls and CKD patients. The degree of variance using Coefficient of Determination (CD) was more than twice that of Cockcroft and Gault in disease state (16.3% vs 6.3%) and its predictive accuracy in health is even worse compared with Cockcroft and Gault (49.3% vs 9.9%). As previously published, the extreme of serum creatinine concentration did not appear to have an important effect on correlation between the predicted and measured creatinine clearance as judged from regression parameters.
It is therefore concluded that Cockcroft and Gault formula still remains the best in our setting and can be used to evaluate stable CKD patients and normal individuals who require the determination of creatinine clearance. Another major advantage of this formula is its simplicity and the relative ease to recall. Hence, its use is emphasized particularly in homogenous African Black populations.
References | | |
1. | Bray JJ, Cragg PA, Anthony DC, Roland GM, Douglas WT (eds). Kidney Water and Electrolytes in Lectures Notes on Human Physiology (Chapter 13) Blackwell Scientific Publications Oxford, 1986; 464-8. |
2. | National Kidney Foundation: K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease. Evaluation, Classification and Stratification. Am J Kidney Dis 2002;[Suppl]:S1-266. |
3. | Payne RB. Creatinine Clearance: a redundant clinical investigation. Ann Clin Biochem 1986; 23(.3):243-50. |
4. | Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130(6):461-470. |
5. | Sanusi AA, Akinsola A, Ajayi AA. Creatinine clearance estimation from serum creatinine values: Evaluation and comparison of five prediction formulas in Nigerian patients. Afr J Med Sci 2000;29:7-11. |
6. | Cockcroft DW, Gault MW. Prediction of Creatinine Clearance from serum creatinine. Nephron 1976;16(1):31-41. |
7. | Rule AD, Larson TS, Bergstrath EJ, Slezak JM, Jacobsen SJ, Cosio FG. Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med 2004;141(12):929-37. |
8. | Poggio ED, Wang X, Greene T Van Lente F, Hall PM. Performance of the Modification of diet in renal disease and Cockcroft-Gault equations in the estimation of GFR in health and in chronic kidney disease. J Am Soc Nephrol 2005;16(2):459-66. |
9. | Froissart M, Rossert J, Jacquort C, Paillard M, Houillier P. Predictive performance of the modification of diet in renal disease and CockcroftGault equations for estimating renal function. J Am Soc Nephrol 2005;16(3):763-73. |
10. | Jelliffe RW. Estimation of Creatinine Clearance when urine cannot be collected. Lancet 1971; 1(7706):975-76. |
11. | Jelliffe RW, Creatinine clearance: bedside estimate. Ann Intern Med 173;79:604-5. |
12. | Mawer EG, Lucas SB, Knowles BR, Stirland RM. Computer assisted prescribing of kanamycin for patients with renal insufficiency. Lancet 1972;1(7740):12-5. |
13. | Hull JH, Hak LJ, Koch GG, et al. Influence of range of renal function and liver disease on predictability of creatinine clearance. Clin Pharmac Ther 1981;29:516-21. |
14. | Gates GF. Creatinine clearance estimation from serum creatinine values: an analysis of three mathematical models of glomerular function. Am J Kidney Dis 1985;5(3):199-205. |
**
**
**Correspondence Address**: Sanusi Abubakr Abefe Department of Medicine, Obafemi Awolowo University Teaching Hospitals Complex, PMB 5538, Ile-Ife, Osun State, Post Code 220001 Nigeria
| **Check** |
**PMID:** 19112223
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2] |